A particular interest in modern drug discovery is the development of novel low molecular weight orally-bioavailable drugs that work by binding to RNA. RNA, which serves as a messenger between DNA and proteins, was thought to be an entirely flexible molecule without significant structural complexity. Recent studies have revealed a surprising intricacy in RNA structure. RNA has a structural complexity rivaling proteins, rather than simple motifs like DNA. Genome sequencing reveals both the sequences of the proteins and the mRNAs that encode them. Since proteins are synthesized using an RNA template, such proteins can be inhibited by preventing their production in the first place by interfering with the translation of the mRNA. Since both proteins and the RNAs are potential drug targeting sites, the number of targets revealed from genome sequencing efforts is effectively doubled. These observations unlock a new world of opportunities for the pharmaceutical industry to target RNA with small molecules.
Classical drug discovery has focused on proteins as targets for intervention. Proteins can be extremely difficult to isolate and purify in the appropriate form for use in assays for drug screening. Many proteins require post-translational modifications that occur only in specific cell types under specific conditions. Proteins fold into globular domains with hydrophobic cores and hydrophilic and charged groups on the surface. Multiple subunits frequently form complexes, which may be required for a valid drug screen. Membrane proteins usually need to be embedded in a membrane to retain their proper shape. The smallest practical unit of a protein that can be used in drug screening is a globular domain. The notion of removing a single alpha helix or turn of a beta sheet and using it in a drug screen is not practical, since only the intact protein may have the appropriate 3-dimensional shape for drug binding. Preparation of biologically active proteins for screening is a major limitation in classical high throughput screening. Quite often the limiting reagent in high throughput screening efforts is a biologically active form of a protein which can also be quite expensive.
For screening to discover compounds that bind RNA targets, the classic approaches used for proteins can be superceded with new approaches. All RNAs are essentially equivalent in their solubility, ease of synthesis or use in assays. The physical properties of RNAs are independent of the protein they encode. They may be readily prepared in large quantity through either chemical or enzymatic synthesis and are not extensively modified in vivo. With RNA, the smallest practical unit for drug binding is the functional subdomain. A functional subdomain in RNA is a fragment that, when removed from the larger RNA and studied in isolation, retains its biologically relevant shape and protein or RNA-binding properties. The size and composition of RNA functional subdomains make them accessible by enzymatic or chemical synthesis. The structural biology community has developed significant experience in identification of functional RNA subdomains in order to facilitate structural studies by techniques such as NMR spectroscopy. For example, small analogs of the decoding region of 16S rRNA (the A-site) have been identified as containing only the essential region, and have been shown to bind antibiotics in the same fashion as the intact ribosome.
The binding sites on RNA are hydrophilic and relatively open as compared to proteins. The potential for small molecule recognition based on shape is enhanced by the deformability of RNA. The binding of molecules to specific RNA targets can be determined by global conformation and the distribution of charged, aromatic, and hydrogen bonding groups off of a relatively rigid scaffold. Properly placed positive charges are believed to be important, since long-range electrostatic interactions can be used to steer molecules into a binding pocket with the proper orientation. In structures where nucleobases are exposed, stacking interactions with aromatic functional groups may contribute to the binding interaction. The major groove of RNA provides many sites for specific hydrogen bonding with a ligand. These include the aromatic N7 nitrogen atoms of adenosine and guanosine, the O4 and O6 oxygen atoms of uridine and guanosine, and the amines of adenosine and cytidine. The rich structural and sequence diversity of RNA suggests to us that ligands can be created with high affinity and specificity for their target.
Although our understanding of RNA structure and folding, as well as the modes in which RNA is recognized by other ligands, is far from being comprehensive, significant progress has been made in the last decade (Chow, C. S.; Bogdan, F. M., Chem. Rev., 1997, 97, 1489, Wallis, M. G.; Schroeder, R., Prog. Biophys. Molec. Biol. 1997, 67, 141). Despite the central role RNA plays in the replication of bacteria, drugs that target these pivotal RNA sites of these pathogens are scarce. The increasing problem of bacterial resistance to antibiotics makes the search for novel RNA binders of crucial importance.
Certain small molecules can bind and block essential functions of RNA. Examples of such molecules include the aminoglycoside antibiotics and drugs such as erythromycin which binds to bacterial rRNA and releases peptidyl-tRNA and mRNA. Aminoglycoside antibiotics have long been known to bind RNA. They exert their antibacterial effects by binding to specific target sites in the bacterial ribosome. For the structurally related antibiotics neamine, ribostamycin, neomycin B, and paromomycin, the binding site has been localized to the A-site of the prokaryotic 16S ribosomal decoding region RNA (Moazed, D.; Noller, H. F., Nature, 1987, 327, 389). Binding of aminoglycosides to this RNA target interferes with the fidelity of mRNA translation and results in miscoding and truncation, leading ultimately to bacterial cell death (Alper, P. B.; Hendrix, M.; Sears, P.; Wong, C., J. Am. Chem. Soc., 1998, 120, 1965).
There is a need in the art for new chemical entities that work against bacteria with broad-spectrum activity. Perhaps the biggest challenge in discovering RNA-binding antibacterial drugs is identifying vital structures common to bacteria that can be disabled by small molecule drug binding. A challenge in targeting RNA with small molecules is to develop a chemical strategy which recognizes specific shapes of RNA. There are three sets of data that provide hints on how to do this: natural protein interactions with RNA, natural product antibiotics that bind RNA, and man-made RNAs (aptamers) that bind proteins and other molecules. Each data set, however, provides different insights to the problem.
Several classes of drugs obtained from natural sources have been shown to work by binding to RNA or RNA/protein complexes. These include three different structural classes of antibiotics: thiostreptone, the aminoglycoside family and the macrolide family of antibiotics. These examples provide powerful clues to how small molecules and targets might be selected. Nature has selected RNA targets in the ribosome, one of the most ancient and conserved targets in bacteria. Since antibacterial drugs are desired to be potent and have broad-spectrum activity these ancient processes fundamental to all bacterial life represent attractive targets. The closer we get to ancient conserved functions the more likely we are to find broadly conserved RNA shapes. It is important to also consider the shape of the equivalent structure in humans, since bacteria were unlikely to have considered the therapeutic index of their RNAs while evolving them.
A large number of natural antibiotics exist, these include the aminoglycosides, kirromycin, neomycin, paromomycin, thiostrepton, and many others. They are very potent, bactericidal compounds that bind RNA of the small ribosomal subunit. The bactericidal action is mediated by binding to the bacterial RNA in a fashion that leads to misreading of the genetic code. Misreading of the code during translation of integral membrane proteins is thought to produce abnormal proteins that compromise the barrier properties of the bacterial membrane.
Antibiotics are chemical substances produced by various species of microorganisms (bacteria, fungi, actinomycetes) that suppress the growth of other microorganisms and may eventually destroy them. However, common usage often extends the term antibiotics to include synthetic antibacterial agents, such as the sulfonamides, and quinolines, that are not products of microbes. The number of antibiotics that have been identified now extends into the hundreds, and many of these have been developed to the stage where they are of value in the therapy of infectious diseases. Antibiotics differ markedly in physical, chemical, and pharmacological properties, antibacterial spectra, and mechanisms of action. In recent years, knowledge of molecular mechanisms of bacterial, fungal, and viral replication has greatly facilitated rational development of compounds that can interfere with the life cycles of these microorganisms.
At least 30% of all hospitalized patients now receive one or more courses of therapy with antibiotics, and millions of potentially fatal infections have been cured. At the same time, these pharmaceutical agents have become among the most misused of those available to the practicing physician. One result of widespread use of antimicrobial agents has been the emergence of antibiotic-resistant pathogens, which in turn has created an ever-increasing need for new drugs. Many of these agents have also contributed significantly to the rising costs of medical care.
When the antimicrobial activity of a new agent is first tested a pattern of sensitivity and resistance is usually defined. Unfortunately, this spectrum of activity can subsequently change to a remarkable degree, because microorganisms have evolved the array of ingenious alterations discussed above that allow them to survive in the presence of antibiotics. The mechanism of drug resistance varies from microorganism to microorganism and from drug to drug.
The development of resistance to antibiotics usually involves a stable genetic change, heritable from generation to generation. Any of the mechanisms that result in alteration of bacterial genetic composition can operate. While mutation is frequently the cause, resistance to antimicrobial agents may be acquired through transfer of genetic material from one bacterium to another by transduction, transformation or conjugation.
One group has prepared and assayed 6′-N-methyl and 6′-N-ethyl derivatives of amikacin (Umezawa, et al., Journal of Antiboiotics, 1975, 28(6), 483-485). They showed that these derivatives were hardly affected by the 6′-N-acetyltransferase which has been shown to inactivate amikacin.
For the foregoing reasons, there is a need for new chemical entities that possess antimicrobial activity. Further, in order to accelerate the drug discovery process, new methods for synthesizing aminoglycoside antibiotics are needed to provide an array of compounds that are potentially new drugs for the treatment microbial infections.